Fail Fast, Learn Faster
-“We can’t make any mistake or else we’re doomed!”
I bet you’ve heard that at some point in your career path… This is what is called a fear-driven culture. A culture where no one is willing to take risks… not even small ones. This is directly related to how fast a company can adapt to market changes in our volatile world. A company that is not willing to take risks will never consider changing its processes in any way. Any change is considered a big risk and would be promptly avoided. There is no place for innovation within those companies.
Don’t get me wrong… I’m not saying that there isn’t any successful company with this type of culture. But at some point in time, they will get stagnated. Frozen by the fear of change…
A fail-safe culture is exactly the opposite. Instead of avoiding mistakes, the companies actually embrace them through what is called experimentation. Roughly speaking, in an experiment we define a hypothesis and test it in a controlled environment. It will either achieve expected results or fail and we will have to study the data, understand what happened, and run another experiment formulating a new hypothesis. This cycle goes on until we are satisfied with the results or the knowledge we’ve gained, which should guide our future directions. The goal is to learn as much as we can, as faster as possible so that we can make the best decisions!
One tool that we can use to guide our experimentation process is the PopcornFlow which is a method that fosters continuous change and evolution and consists of two parts: a decision cycle and a set of principles. Here is an interview with Claudio Perrone, creator of the method, where he gives an overview of it.
As Thomas Edison once said, he has not failed 10,000 times, he’s just acquired knowledge. 3M Post-it, penicillin, Coca Cola’s New Coke are all results of mistakes. Innovation comes from iterative learning processes in fail-safe creative environments. According to Soichiro Honda, “Success is 99% failure”. Mistakes have the power to show us different and sometimes unexpected possible ways which could be even better than the original planned path. If leaders do not allow failures, they limit the companies creative capacity and kill innovation.
People who take small risks, fail, and learn with them, will know how to avoid them in the future. Experiment & Learn Rapidly is one of the principles of Modern Agile and it is directly related to the innovation process! Some companies even celebrate failure as they are learning from it. That is what Jurgen Appelo suggests with the Management 3.0 Celebration Grid which is a diagram that helps us visualize the outcome of our experiments.
Some things to keep in mind while experimenting:
1- Avoid large risks, aim for small risks. People who take large risks and fail will be crushed by their mistakes. Those who take this approach and luckily make it to the top without failing will just become overconfident, which will lead to even larger risks… One way to reduce the risk is having better control over the experiment variables. A good definition of the problem may help us achieve that.
2- It is important to know how and what to measure and control possible impacts and take actions to mitigate them when the experiment fails. In software development, there are several tools and practices that help us accomplish that such as automated tests, feature toggles, CI and CD tools, A/B testings, control metrics, feedback loops, etc…
3- Set your hypothesis, define a baseline, the metrics and data that you’ll monitor during your experiment. Remember that this is a learning process. Whatever the result is, we should always learn something from it.
Don’t be discouraged by a failure. Understand what went wrong, learn with it and get ready for the next time! This is part of the continuous improvement process and it is something that everyone and every company should pursue.
So, do yourself a favor and go make some mistakes! :)